Adsorption Behavior of Heavy Metal Ions from Aqueous Solution by

Adsorption Behavior of Heavy Metal Ions from Aqueous Solution bySoy Protein Hollow MicrospheresDagang Liu,* Zehui Li, Wei Li, Zhengrong Zhong, Jianqiang Xu, Jinjing Ren, and Zhongshi Ma

Department of Chemistry, Nanjing University of Information Science and Technology, Nanjing, 210044, China

ABSTRACT: Heavy metals have become ecotoxicological hazards owing to their tendency to not degrade but accumulate in thevital organs of biological bodies. Biosorption is now an efficient method to purify industrial wastewater containing toxic heavymetal ions by using biomass as sorbents. In this work, mimicking the fabricating process of “Tofu”, soy protein was heat-denatured and transformed into soy protein hollow microspheres (SPMs) with a diameter of about 4−45 μm, which were thenused as biosorbents to remove heavy metal ions in the water system. The trace amount of remaining metal ions was tested byatomic absorption spectroscopy, and the sorption kinetic and isotherm models were calculated and set up to describe theadsorption behavior. The results showed adsorption capacities of Zn(II), Cr(III), Cd(II), Cu(II), Pb(II), and Ni(II) by SPMs at70 °C of 254.95, 52.94, 120.83, 115.01, 235.56, and 177.11 mg/g, respectively, which are much higher than that of many othernatural polymeric sorbents. The pseudofirst-order kinetic model and Freundlich isotherm model were well correlated to theexperimental data. Overall, SPMs were efficient sorbents for binding heavy metal ions, and their sorption capacities weredependent on factors like denaturation content, temperature, time, pH, and initial ionic concentration.

1. INTRODUCTIONWater pollution caused by heavy metal ions had been attractinggreat attention in the modern industrial society, because heavymetal ions even in trace amount are hazardous to human healthand the ecosystem and are difficult to remove even at lowconcentration.1,2 Various methods for the removal of toxicmetals from an aqueous system have been developed for a longperiod,3 such as ion exchange, reverse osmosis, membranefiltration, complexation precipitation, and adsorption.4 Nor-mally, adsorption is a highly effective, environmentally friendly,inexpensive, and easy way to operate among thosephysicochemical treatment processes,5 especially method ofbiosorption. Biosorption was used to describe the ability ofbiological materials like biopolymers to concentrate or removeorganic or inorganic pollutants like heavy metals by the passivebinding to nonliving biomass from an aqueous solution.6 Manymacromolecules extracted from a biological body, such aschitosan,7 cellulose,8,9 starch,10 alginate,11 and so on, have greatpotentials of adsorbing heavy metal ions from wastewaterbecause they bear high contents of hydroxyl, amino, and otheractive functional groups on molecular chains. However, theadsorption capacity of these unmodified biopolymers waslimited. To improve adsorption performance, micro- andnanotechnology applied on biopolymer absorbents were stillunder development.Recently, biopolymer microspheres12,13 have been paid

attention due to the good performance of such microspheres,such as low toxicity, good biocompatibility, low cost,minimization of secondary wastes and biodegradability, largesurface area, and high mechanical stability. Among the mostused biopolymers, modified chitosan beads showed a highcapability of absorbing Cu(II) ions up to 64.62 mg/g.14 Yang etal.15 prepared neutral starch microspheres cross-linked byepichlorohydrin using an inverse microemulsion method. Thereported anionic starch microspheres with average diameter of75 μm had good sphericity, fine dispersibility, and good ability

of sorption heavy metals. The adsorbed amounts of starchmicrospheres were 75.99−83.33 mg/g for Cu(II) and 62.40−66.67 mg/g for Pb(II) ions, respectively. Meanwhile, chitin andalginate microspheres have also been explored for heavy metalion treatment.Soy protein is composed of a mixture of globular proteins,

which can be divided into 2S, 7S, 11S, or 15S fractions on thebasis of molecular weight and sedimentation coefficient.16

Among those fractions, 7S and 11S globulin are two mainglobular proteins, with amounts up to 37% and 31%,respectively.17 Globular proteins are composed of segmentsof polypeptides connected with hydrogen bonds, electrostaticinteractions, disulfide bonds, and hydrophobic interactions.18

Once exposed to pH, ionic potential, heat, and other factors,conformational changes of unfolding globular proteins througha physical/chemical process would cause the denaturation ofnative globular proteins, i.e., converting into unfoldedpolypeptide chains, which are connected with interchangingof disulfide bonds. It has been widely reported that soy proteincan be precipitated by adding many Ca(II) or Mg(II) ions inorder to make a food “Tofu” as a result of metal ion-inducedassociation. On the basis of the “salt-out” theory, we coulddeduce that heavy metal ions can be combined to soy proteinand then taken away from an aqueous system. Soy proteinhydrogel modified by ethylenediaminetetraacetic acid dianhy-dride was also reported to have good metal-chelatingproperties.19,20 In this work, soy protein microsphere wouldbe fabricated and used to uptake heavy metals ions;simultaneously, various parameters like contact time, ionicconcentration, and adsorption temperature were investigated to

Received: April 7, 2013Revised: July 8, 2013Accepted: July 20, 2013Published: July 20, 2013

Article

pubs.acs.org/IECR

© 2013 American Chemical Society 11036 dx.doi.org/10.1021/ie401092f | Ind. Eng. Chem. Res. 2013, 52, 11036−11044

pubs.acs.org/IECR

determine adsorption behavior of soy protein hollow micro-spheres (SPMs) in an aqueous system. We aim to set up ameaningful sorption model about this novel biosorbent forfuture application of heavy metal environmental treatments.

2. MATERIALS AND METHODS

2.1. Materials. Commercial soy protein isolate (SPI)purchased from Dupont-Yunmeng Protein Technology Co.Ltd. (Yunmeng, China) was used as pure soy protein. Theoriginal moisture and protein content of SPI (dry basis) were5.0% and 92.3%, respectively. Cadmium(II) nitrate tetrahy-drate, lead(II) nitrate, nickel(II) nitrate hexahydrate, cupric(II)sulfate anhydrous, chromium(III) chloride hexahydrate, andzinc(II) sulfate heptahydrate were purchased from SinopharmChemical Reagent Co., Ltd. (Shanghai, China). All thesechemicals were analytical grade and used as received withoutfurther purification.2.2. Preparation of SPMs and Removal of Heavy Metal

Ions. A target amount of SPI was vacuum-dried at 100 °C for 2h, and then, 1.0 g of dried SPI was poured into a flaskcontaining 200 mL of deionized water to prepare soy proteinslurry under mechanical agitation at 50 rpm for half an hour at70 °C.Heavy metal ion adsorption experiments were then

performed according to the batch method. A target amountof zinc sulfate dehumidified at 70 °C was added into theprepared soy protein suspension, thus composing simulatedwastewater with different concentrations of Zn(II). Adsorptionof Zn(II) ions by soy protein was achieved at 70 °C undermechanical agitation at 50 rpm for 4 h. After adsorption, soyprotein binding Zn(II) ions were separated from the mixedsuspension by centrifugation at 12 000 rpm (HC-2061centrifuge, Anhui USTC Zonkia Scientific Instruments Co.,Ltd.). Finally, the supernatant was diluted for concentrationanalysis using atomic absorption spectroscopy (AAS, 3510,Shanghai branch, Agilent technologies Inc.). Other ions likeCd(II), Pb(II), Ni(II), Cu(II), and Cr(III) ions with differentconcentrations were adsorbed, separated, and analyzed in thesame way as Zn(II). In addition, Cd(II), Zn(II), and Cr(III)ions were poured into soy protein suspension and stirred for 4h at 30, 40, 50, 60, 70, and 80 °C, respectively. Aftercentrifugation, the separated supernatants were subsequentlyused for AAS testing to analyze the adsorption capacity ofheavy metals by soy protein under varying temperature.Contact time dependent adsorption experiments were carriedout as follows. A target amount of dehumidified zinc sulfate waspoured into soy protein aqueous suspension (2.5 g/L) at 70°C. During the adsorption process, a 25 mL suspension wassampled by pipet (25 mL) at particular intervals, and then, thecorresponding supernatant was immediately centrifuged,collected, and diluted for concentration analysis by AAS.Similarly, the dependence of contact time on adsorption ofother ions like Cd(II), Pb(II), Ni(II), Cu(II), and Cr(III) wastested. During our experiments, we tested the pH of solutionsbefore and after adsorption and found that pH varied within0.05, so the influence could be neglected. Furthermore, noadditional competitive ions from buffer solution would affectthe sorption amount and sorption behavior of heavy metal ions.Therefore, without any buffer solution, nitric acid (0.1 M) wasdropped into soy protein suspension containing heavy metalsalts to adjust systemic pH at 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 5.0,respectively; thus, pH dependent adsorption of heavy metals

was tested by AAS. Adsorption capacity of soy protein qt (mg/g) above-mentioned was expressed as the following equation:

=− ×

qm V C

mt1 i i

(1)

where Ci represents the final metal ion concentrations of thesolution after the adsorption processes (mg/L), Vi representsthe final solution volume (L), m represents adsorbent mass (g),and m1 represents the initial metal ion mass (mg).

2.3. Morphological Characterization of SPMs. Theabove obtained sediment after centrifugation was baked at 40°C in a drying oven for 12 h. The obtained baking “cakes” werecut into small pieces, one of which was further sintered at 350°C for 1 h, and then, the baking “cake” and calcinated “cake”piece were coated with gold for scanning electron microscopy(SEM, Hitachi S-3600N VP SEM, Hitaschi, Japan) observationat 20 kV. The specific surface areas of sorbents were measuredby the nitrogen adsorption/desorption isotherm method atliquid nitrogen temperature using the Automated Gas Sorption(Autosorb-iQ-AG-MP, Quantachrome Co., U.S.A.). TheBrunauer−Emmett−Teller (BET) model was applied tocalculate the apparent surface area.

3. RESULTS AND DISCUSSION3.1. Appearance and Morphology of SPMs. The

Brunauer−Emmett−Teller (BET) surface area of soy protein

and thermally denatured soy protein was 6.20 and 9.76 m2/g,respectively, indicating a higher surface area of the sorbent.Photographs of baking “cake” were displayed in Figure 1, inwhich panels a−f represent SPMs and SPMs adsorbed Cd(II),Cu(II), Pb(II), Ni(II), and Zn(II), respectively. Soy proteinwas easily thermo-heated into weak gel (Figure 1a). Whenheavy metal ions was adsorbed into soy protein, the gel wassolidified and could be baked into “cake” which had a solidform and special color originated from the nature of ions, e.g.,blue cake with Cu(II) ion. Interestingly, the “cake” containingPb(II) and Ni(II) ions is very strong and likely solidified from agelation sediments; that is, these two heavy metal ions weretightly bound onto or into soy protein polypeptides. In the caseof other “cakes” with Cu(II), Zn(II), and Cd(II) ions, manycracks exhibited on their surfaces. It is known that soy proteincan be heat-denatured, which means unfolding chains andfunctional groups such as disulfide groups and hydrophobic

Figure 1. Photographs of SPMs (a) and SPMs binding Cd(II) (b),Cu(II) (c), Pb(II) (d), Ni(II) (e), and Zn(II) (f) ions, respectively.

Industrial & Engineering Chemistry Research Article

dx.doi.org/10.1021/ie401092f | Ind. Eng. Chem. Res. 2013, 52, 11036−1104411037

groups become exposed and immediately interact with eachother, leading to reversible polypeptide aggregation andformation of gelation network during the heating process.When denatured protein encountered heavy metal ions, activesites in the hydrophilic region were exposed to chelate ions andoxidation reactions simultaneously would break the intermo-lecular disulfide bonding bridge via thiol-disulfide inter-change.21 That is why Pb(II) ion is most known as a breakerof disulfide bonding in killing blood protein. Thus, it could be

deduced that heavy metal ions played an important role indenaturation, aggregation (especial disulfide bridges), networkformation, and gel stiffening of soy protein. Adsorption ofheavy metal ions was an irreversible denaturation process of soyprotein.Micrographs of soy protein adsorbed heavy metal ions and its

deposits sintered at 350 °C were shown in Figure 2a, b (Cd2+);c, d (Cr3+); e, f (Cu2+); h, i (Ni2+); j, k (Pb2+); l, m (Zn2+),respectively. As it can be seen, denatured soy protein positively

Figure 2. SEM micrographs of SPM adsorbed Cd(II) (a), Cr(III) (c),Cu(II) (e), Ni(II) (h), Pb(II) (j), and Zn(II) (l) and sinter of SPMadsorbed Cd(II) (b), Cr(III) (d), Cu(II) (f), Ni(II) (i), Pb(II) (k),and Zn(II) (m), calcinated at 350 °C.

Figure 3. Effects of pH on adsorption of metal ions by SPMs (initialconcentration of the heavy metals: 200 mg/L; stirring: 150 rpm; pH:2.0−5.0; temperature: 20 °C; adsorption time: 4 h).

Figure 4. Dependence of adsorption capacity of SPMs on the initialconcentration of Cd(II), Cr(III), Cu(II), Pb(II), Ni(II), and Zn(II)ions (initial concentration of the heavy metals: 0−0.045 mol/L;stirring: 150 rpm; pH: 5.5; temperature: 20 °C; adsorption time: 4 h).

Table 1. Adsorption Capacities (mg/g and mmol/g) ofVarious Heavy Metal Ions by SPMs at 50 and 70 °C

maximum adsorptioncapacity at 50 °C

maximum adsorptioncapacity at 70 °C

metal ions mg/g mmol/g mg/g mmol/g

Zn(II) 174.71 2.69 254.95 3.92Cr(III) 42.60 0.82 52.94 1.02Pb(II) 169.18 0.82 235.56 1.14Cd(II) 106.37 0.94 120.83 1.08Cu(II) 92.43 1.45 115.01 1.80Ni(II) 130.05 2.22 177.11 3.02



existed in the form of microspheres with a diameter range of 4−45 μm. In most cases, the microspheres collapsed during thebaking and sintering process. As can be observed from thecross-section of soy protein sintered in Figure 2i, microsphereswere stacked together and presented a hollow structure. Manysmall particles occurred on the surface of SPMs as observedfrom Figure 2a−f, which were the heavy metal salts with a sizeof 5−50 μm. Some ionic salts were thought to be combined inthe shell of microspheres instead of the outer surface, whichcould be revealed from Figure 2h−m. It was thought thatadsorption of heavy metal ions by SPMs contributed to the

driving forces of electrostatic interactions, chelation, ionicbonding, etc. For instance, heavy metal ions like Zn(II), Ni(II),and Pb(II) could chemically bond to thiol or other groups ofdenatured protein leading to these ions binding the soypolypeptide network and penetrating into the microsphereshell.19,20 Therefore, it can be concluded that plenty ofhydrophilic active sites located on/in the denatured SPMsplayed an important role in adsorption of heavy metals.

3.2. Sorption Behavior of SPMs. 3.2.1. Effects of pH ofSuspension. pH is known as an important factor affecting theassociation degree of metal ions on the biomass surface withmultifunctional groups.22 When the systemic pH of suspensionwas adjusted between 2.0 and 7.0 by using HNO3, the sorption

Figure 5. Effects of contact time on the adsorption capacity of heavy metal ions for SPMs at 70 °C (a) and 50 °C (b) (initial concentration of theheavy metals: 200 mg/L; stirring: 150 rpm; pH: 5.5; temperature: 70 and 50 °C; adsorption time: 0−600 min).

Figure 6. Effects of temperature on the adsorption capacity of Cd(II),Zn(II), and Cr(III) ions for SPMs (initial concentration of the heavymetals: 200 mg/L; stirring: 150 rpm; pH: 5.5; temperature: 30−80 °C;adsorption time: 4 h).

Table 2. Freundlich and Langmuir Isotherm Constants for Metal Ions Binding to SPMs

Langmuir Freundlich

metal ions measured qmax (mg/g) Qmax (mg/g) b (L/mg) R2 1/n KF (mg/g) R2

Cd(II) 120.83 −10000.0 −0.0084 0.0011 1.0517 0.061 0.9441Cr(III) 52.94 250.0 0.4938 0.6189 0.8953 0.192 0.9872Cu(II) 115.01 204.1 0.4414 0.9665 0.6807 0.539 0.9756Ni(II) 177.11 1428.6 0.1321 0.2362 0.9052 0.320 0.9758Pb(II) 235.56 −1111.1 −0.0833 0.0842 1.1775 0.030 0.8995Zn(II) 254.95 277.8 7.2000 0.9989 0.1225 102.849 0.9173

Figure 7. log (qe−qt) as a function of time for Cd(II), Cr(III), Cu(II),Ni(II), Pb(II), and Zn(II) ions, respectively.



capacities varied with pH and are displayed in Figure 3. It canbe observed that the absorption of these heavy metals arehighly pH dependent. The adsorption of Zn(II), Cr(III),Cd(II), Cu(II), Pb(II), and Ni(II) increased with pH increasingfrom 2 to 4 and decreased with pH increasing from 4 to 5. Thehighest value is at pH of 4.0 or 4.5. The facts were thatdissociation degree of functional groups from sorbent surfaceincreased with rising pH and, in consequence, ionic interactionsincreased; e.g, −NH2 of soy protein can chelate with metal ionsand also with H+, so there is a equilibrium between them.Consequently, the concentration of H+ could affect theadsorption capacity of the active groups on soy proteinmicrospheres. As the pH value was reduced, the equilibriumshifted toward the direction to generate R−NH3

+, thus leadingto dissociation of metal ions. It is well-known that theisoelectric point of soy protein is at about 4−5; in other words,the solubility is at a minimum, charges of protein areneutralized, and electrostatic repulsive forces among groupshave disappeared. Our results indicated that adsorption capacitywas consistent with the isoelectric point of soy protein anddramatically changed with the solubility of soy protein whichvaried with pH. Since secondary hydrolysis processes tookplace for those heavy metal salts, a slow decrease of absorptionefficiency is observed as pH values were higher than 4.5 or 5.0.Therefore, an initial pH at 4.5 or 5.0 was considered as anoptimum sorption condition to remove heavy loading metalions.3.2.2. Effects of Initial Concentration of Metal Ions. Figure

4 shows the effects of initial ionic concentration of heavy metalions on adsorption captivity of SPMs. The adsorption capacitiesincreased with the increase of concentration of heavy metal ionsand gradually achieved a maximum adsorption value, which isnoted as the maximum adsorption captivity. The fact is that atlow concentration the ratio of available binding sites to the totalmetal ions was high and all metal ions could be bound to theactive sites of microspheres, whereas at high concentrations, theratio was lower and consequently the binding was dependenton the initial concentrations. As listed in Table 1, the maximumadsorption capacities of Cr(III), Cu(II), Cd(II), Ni(II), Pb(II),and Zn(II) ions by AAS at 70 °C were 52.94, 115.01, 120.83,177.11, 235.56, and 254.95 mg/g or 1.02, 1.80, 1.08, 3.02, 1.14,and 3.92 mmol/g, respectively, and decreased in the order:Zn(II) > Ni(II) > Cu(II) > Pb(II) > Cd(II) > Cr(III). In orderto understand the level of adsorption degree, we made a

Figure 8. t/qt as a function of time for Cd(II), Cr(III), Cu(II), Ni(II),Pb(II), and Zn(II) ions, respectively.

Table

3.Kinetic

Param

etersof

RegressionEqu

ations

ofSorption

Mod

elat

70°C

pseudofirst-order

pseudosecond-order

Elovichequatio

nMorries-Weber

equatio

n

breakpoint

metalions

measuredq e

R2

k 1calculated

q eR2

k 1calculated

q eα(m

g·g−

1 ·min

−1 )

β(g·m

g−1 )

R2

t 1(m

in)

q 1(%

)R12

R22

Cd(II)

114.75

0.9999

0.0124

120.23

0.9906

0.0006

135.32

84.7338

0.0400

0.8884

240

95.41

0.9838

0.8938

Cr(III)

51.74

1.0000

0.0192

55.21

0.9994

0.0015

55.59

17.6036

0.1308

0.8576

240

96.88

0.9681

0.9988

Cu(II)

117.09

0.9999

0.0166

123.22

0.9977

0.0006

129.53

29.7230

0.0458

0.9285

300

98.70

0.9744

0.9987

Ni(II)

150.28

0.9999

0.011

175.39

0.9903

0.0004

207.04

390.2682

0.0222

0.9511

300

95.00

0.9980

0.8784

Pb(II)

232.84

0.9999

0.0072

233.71

0.9059

0.0003

325.73

1224.7328

0.0133

0.8948

240

95.88

0.9950

0.7976

Zn(II)

250.86

0.9999

0.0113

254.68

0.9825

0.0003

294.12

149.0023

0.0198

0.7747

240

99.05

0.9581

0.7194



comparison of soy protein with other effective biopolymeradsorbents, such as chitosan. Chitosan was reported to havesaturated adsorption capacities of 0.5, 1.5−2.5, 0.6−1.3, 2.0, 0.2,and 0.94 mmol/g corresponding to Cr(III),23 Cu(II),24

Cd(II),25 Ni(II),26 Pb(II),27 and Zn(II),28 respectively. It iseasy to find that SPMs had a higher capability of binding heavymetals than chitosan except the case of comparable adsorptionfor Cu(II) and Cd(II).3.2.3. Effects of Contact Time and Temperature. The time

profiles of metal ions sorption by SPMs at 70 and 50 °C arepresented in Figure 5a,b, respectively. The amount of metalions adsorbed did not increase with contact time until theplateau, implying equilibrium was reached. The amount ofmetal ions adsorbed rapidly in the initial stage was attributed tothe availability of binding sites of the microspheres, but as timewent by, the sorption slowed down before reaching equilibriumat about 4 h at 70 °C. The equilibrium values of Cd(II) 114.75mg/g, Cr(III) 51.74 mg/g, Cu(II) 117.09 mg/g, Ni(II) 150.28mg/g, Pb(II) 232.84 mg/g, and Zn(II) 250.86 mg/g could beachieved at a high level of adsorption efficiency. The timeprofile at 50 °C had a similar trend to reach equilibrium atabout 4 h, and the equilibrium values were Cd(II) 106.40 mg/g, Cr(III) 42.27 mg/g, Cu(II) 77.59 mg/g, Ni(II) 113.46 mg/g,Pb(II) 180.52 mg/g, and Zn(II) 175.13 mg/g. Maximumsorption capacities of heavy metal ions by SPMs at 70 and 50°C are summarized in Table 1, respectively. Obviously, theequilibrium sorption amount of heavy metal ions on SPMs at70 °C was higher than that at 50 °C, which means the bindingmechanism was dependent on the temperature and therebyheating denaturation of soy protein.

When temperature was extended from 30 to 80 °C,temperature profiles of metal ion uptake of three specialcases of Zn(II), Cr(III), and Cd(II) were shown in Figure 6.The adsorption capacities of Cd(II), Cr(III), Cu(II), Ni(II),Pb(II), and Zn(II) at 30 °C were 83.36, 31.43, 65.12, 119.32,128.54, and 122.29 mg/g, respectively, whereas at 80 °C theywere enhanced up to 145.18, 53.39, 121.34, 153.54, 235.34, and253.34 mg/g, respectively. The adsorption capacity wasincreased with increasing temperature from 30 to 80 °C. Asharp elevation occurred between 50 and 70 °C which wascaused by a critical denaturation of soy protein at thistemperature range. Therefore, the active binding sites exposedon the microspheric surface were correlated with the degree ofprotein denaturation, and above 70 °C, the binding process wasalmost completed, thus leading to no significant changes ofionic uptake.

3.3. Sorption Isotherm. Two major isotherm equations,the Langmuir and Freundlich isotherms,29,30 are established tofit the experimental data and describe the isotherm constants ofmetal ion uptake by sorbents. The Langmuir adsorptionisotherm can be expressed as:

= +CQ Q

CQ b

1 1e

e maxe

max (2)

where Ce is the equilibrium concentration of metal ions (mg/L), Qe is the adsorption amount at the equilibrium (mg/g),Qmax is the maximum capacity (mg/g), and b is the Langmuirconstant related to the affinity of binding sites (L/mg).The linear form of the Freundlich isotherm is expressed as:

= +Q Kn

Clog log1

loge F e (3)

where KF is roughly an indicator of the adsorption capacity and1/n is the adsorption intensity. The magnitude of the exponent1/n gives an indication of the favorability of adsorption. Thevalue of 1/n is more than 1, which represents favorableadsorption condition.31

The parameters from the Langmuir and Freundlich modelare calculated and listed in Table 2. The Freundlich equationrepresented a better fit of experimental data than the Langmuirequation for metal ions although it seemed that the model ofZn(II) ions was also suitable for the Langmuir equation. The 1/n value can be used to predict binding affinity of the sorbentstoward metal ions; a smaller value of 1/n implied a strongerinteraction between sorbent and metal ions. As presented inTable 2, the values of 1/n from application of the Freundlichmodel lying between 0 and 1.2 indicated that the metal ionswere favorably adsorbed by the SPMs, and the 1/n values werein the decreasing order of Zn(II) < Cu(II) < Cr(III) < Ni(II) <

Table 4. Pseudofirst-Order and Pseudosecond-Order Kinetic Parameters Obtained from the Linear Fits of Experimental Datafor Cd(II), Cr(III), and Zn(II) at 50 °C

pseudofirst-order pseudosecond-order

metal ions measured qe R2 k1 calculated qe R2 k1 calculated qe

Cd(II) 106.40 0.9999 0.0093 105.51 0.9327 0.0006 180.83Cr(III) 42.27 0.9998 0.0091 42.66 0.9961 0.0016 44.98Cu(II) 77.59 0.9912 0.0088 85.27 0.9989 0.0001 103.09Ni(II) 113.46 0.9990 0.0076 141.64 0.9969 0.0001 163.93Pb(II) 180.52 0.9919 0.0069 190.99 0.9935 0.0001 256.41Zn(II) 175.13 0.9999 0.0097 173.78 0.8928 0.0004 268.10

Table 5. Activation Energy Calculated from the Pseudofirst-Order Kinetic Model Obtained from the Linear Fits ofExperimental Data to the First-Order Rate Equation ForCd(II), Cr(III), and Zn(II) Ions

metalions

temperature(K)

qe(mg/g)

apparent rateconstant k

Ea(kJ/mol)

Cd(II) 323.15 106.37 0.0093 13.11343.15 120.83 0.0124

Cr(III) 323.15 42.60 0.0091 34.40343.15 52.94 0.0192

Cu(II) 323.15 89.07 0.0088 29.26343.15 115.01 0.0166

Ni(II) 323.15 136.95 0.0076 17.04343.15 177.11 0.0110

Pb(II) 323.15 215.17 0.0069 19.62343.15 235.56 0.0072

Zn(II) 323.15 174.71 0.0097 6.85343.15 254.95 0.0113



Cd(II) < Pb(II), suggesting the strongest binding interactionbetween Zn(II) and active sites of the SPMs.3.4. Sorption Kinetics. 3.4.1. Pseudofirst-Order Kinetics

and Pseudosecond-Order Kinetics. Sorption rate is animportant factor for determining the efficiency of a sorptionprocess. Of all adsorption kinetic models, the pseudofirst-orderkinetics and pseudosecond-order kinetics are the most widelyused models to determine the rate constant and the controllingmechanism of the sorption process. The expression of thepseudofirst-order kinetics model is given by following:

= −q

tk q q

d

d( )t

t1 e (4)

The linear form of this model obtained by the integral couldbe expressed as eq 5:

− = −q q qk t

log( ) log2.303te e

1(5)

where k1 is the pseudofirst-order rate constant (min−1), qe is theamount of heavy metal ions adsorbed at equilibrium (mg·g−1),and qt is the amount of the adsorption at any time t (mg·g−1).Such an equation should yield a straight line with interceptequal to logqe and slope equal to −(k1/2.303).As could be observed from the expression above, the

pseudofirst-order kinetics model is based on the assumptionthat adsorption was controlled by diffusion steps, and the rateof adsorption is in direct proportion to the difference value ofequilibrium adsorption capacity and the adsorption capacity atany time t. While the expression of pseudosecond-orderkinetics model is given by following:

= −q

tk q q

d

d( )t

t2 e2

(6)

The linear form of this model obtained by the integral couldbe expressed as eq 7:

= +tq k q

tq

1

t 2 e2

e (7)

where k2 is the pseudosecond-order rate constant(g·mg−1·min−1), qe is the amount of heavy metal ions adsorbedat equilibrium (mg·g−1), and qt is the amount of the adsorptionat any time t (mg·g−1).On the basis of the time profiles above-mentioned, the

pseudofirst-order and pseudosecond-order dynamic modelswere fitted and set up in Figures 7 and 8. The kineticparameters obtained by the sorption of heavy metal ions onSPMs at 70 °C are summarized in Table 3. Linear correlationcoefficients (R2) from the pseudofirst-order dynamic modelwere closer to 1 than those from the pseudosecond-orderdynamic model, and the experimental equilibrium sorptioncapacities determined from the contact time profiles correlatedwell to theoretical equilibrium sorption capacity calculatedusing the pseudofirst-order dynamic model. Thus, it isobviously deduced that the adsorption fits to the pseudofirst-order kinetic model better. Using the same method, thesorption kinetic at the temperature of 50 °C was also describedby the pseudofirst-order and the pseudosecond-order dynamicmodel, respectively. The heavy metal sorption at 50 °C stillfollowed the same model (the pseudofirst-order), which meansthe sorption mechanism does not change with the varyingtemperature at all. As shown in Table 4, the sorption at 50 °Chad a lower sorption rate constant than that at high

temperature (70 °C) as well as less equilibrium uptake.Therefore, the higher temperature was favored for moreeffectively and faster adsorption of heavy metal ions by theSPMs.

3.4.2. Intraparticular Model (Morries-Weber Equation).When absorbates transmit from solution into solid phase likeabsorbents, pore and intraparticle diffusion are often rate-limiting in a batch reactor system. The intraparticle diffusionwas explored by using the following equation suggested byWeber and Morries:32

=q k tt p0.5

(8)

where the parameter qt is the amount adsorbed at time t(mg·g−1), kp is the intraparticle diffusion equation constant(mg·g−1·min−0.5), and t is the time.According to the Weber-Morries model, the plot of qt,

against t0.5, should give a straight line when diffusion plays arole in the sorption rate and should cross the origin ifintraparticle diffusion is the rate determining step.32,33 Theresulted parameters are listed in Table 3; the Weber-Morriesmodel had two linear regions with one break point. The q1%before break point at about 240−300 min was more than 95%,indicating that the adsorption rate of heavy metals are higher inthe beginning owing to the large surface area of the adsorbentavailable for sorption. Then, adsorbate formed a thick layer inthe exterior gradually due to the inter attraction and molecularassociation. This blocked the further adsorption, and the uptakerate was limited during transport from the exterior to theinterior sites of the SPMs.

3.4.3. Elovich Equation. If the process is a chemisorption onhighly heterogeneous sorbents, the sorption kinetics could beinterpret by Elovich equation as follows:34

βαβ

β= +

⎛⎝⎜

⎞⎠⎟q t

1ln( )

1ln( )t (9)

where qt is the sorption capacity (mg·g−1) at time t (min), α isthe initial adsorption rate (mg·g−1·min−1), and β is thedesorption constant (g·mg−1). α and β can be obtained fromthe slopes and intercepts of qt versus ln t plots. The Elovichequation is established if the process is based on diffusion orchemical reaction control. When the adsorption is on anenergetically heterogeneous surface, the parameter α is relatedto the distribution of activation energies, and β is a function ofboth the particle structural−chemical characteristics and solutediffusion coefficient. On the basis of the above Elovich models,the calculated parameters are listed in Table 3. The Elovichmodel shows a preferable linear relationship (R2 > 0.85, exceptZn(II)) and high α, indicating quick and effective sorption.

3.5. Activation Energy of Sorption. The activationenergy can be thought of as the minimum kinetic energyrequired for a particular reaction to overcome the energeticbarrier that the adsorbate ions could be fixed by the adsorbents.The activation energy of the adsorption (Ea: J/mol) can becalculated using the following equation:

= −⎛⎝⎜

⎞⎠⎟

k Tk T

ER T T

ln( )( )

1 12

1

a

1 2 (10)

where k is apparent rate constant, Ea is activation energy, R isgas constant (8.314 J/mol·K), and T is temperature (K).According to the parameters of the pseudofirst-order kineticmodel, activation energies of Zn(II), Cr(III), and Cd(II)



adsorption by SPMs are calculated and summarized in Table 5.As seen from the table, the value of Ea related to adsorption ofZn(II) ion was lower than that of other metal ions, indicatingthat the energetic barrier against the adsorption of Zn(II) ionwas easier to overcome; therefore, the adsorption reaction ofZn(II) was more facile to occur than that of other ions. Cr(III)ion was the most difficult to combine on soy protein among thethree metal ions. Simultaneously, it is easy to find that theequilibrium adsorption capacity of three ions had the oppositeorder with Ea, which provides a reasonable mechanism of whysoy protein had selective adsorption capacity for heavy metalions.

4. CONCLUSIONS

In summary, SPM with diameter ranging from 4 to 45 μm wasprepared by heat-induced denaturation and then utilized asbiosorbent for heavy metal ions. The higher denaturationwould lead to the more hydrophilic active groups exposed onthe outside of microspheres and hence increased the bindingsites for metal ions by increasing temperature and time. Themaximum adsorption capacities of Cr(III), Cu(II), Cd(II),Ni(II), Pb(II), and Zn(II) ions at 70 °C were 52.94, 115.01,120.83, 177.11, 235.56, and 254.95 mg/g or 1.02, 1.80, 1.08,3.02, 1.14, and 3.92 mmol/g, respectively, and decreased in theorder: Zn(II) > Ni(II) > Cu(II) > Pb(II) > Cd(II) > Cr(III).SPM displayed a great advantage over some other metal ionbiosorbents since it has a high sorption capacity. Theadsorption kinetic and isotherm models were calculated andfitted the pseudofirst-order kinetic model and Freundlichmodel, respectively. The selectivity of heavy metal ions bindingonto microspheres resulted from sorption active energy; thus,the maximum adsorption capacity of Zn(II) ion was muchhigher than that of Cd(II) and Cr(III) ions because of thelower energetic barrier. Additionally, adsorption was dependenton pH value, and maximum sorption was reached at about theisoelectric point of soy protein. All in all, SPMs were facile toprepare and easily separated from an aqueous system bycentrifugal settlement, which is anticipated to be widely used inheavy metal treatments in the future.

■ AUTHOR INFORMATION

Corresponding Author*Tel./Fax: +86 2558731090. E-mail: [email protected].

NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTS

The authors are grateful to National Natural ScienceFoundation of China (Nos. 51103073 and 21277073), NaturalScience Foundation of Jiangsu Province (No. BK2011828), andQing Lan Project of Jiangsu Province for financial support.

■ REFERENCES(1) Zhang, Q.; Gao, Y.; Zhai, Y. A.; Liu, F. Q.; Gao, G. Synthesis ofsesbania gum supported dithiocarbamate chelating resin and studieson its adsorption performance for metal ions. Carbohydr. Polym. 2008,73 (2), 359−363.(2) Brower, J. B.; Ryan, R. L.; Pazirandeh, M. Comparison of ion-exchange resins and biosorbents for the removal of heavy metals fromplating factory wastewater. Environ. Sci. Technol. 1997, 31 (10), 2910−2914.

(3) Chang, Y.-C.; Chang, S.-W.; Chen, D.-H. Magnetic chitosannanoparticles: Studies on chitosan binding and adsorption of Co(II)ions. React. Funct. Polym. 2006, 66 (3), 335−341.(4) Zeng, X.; Ruckenstein, E. Cross-linked macroporous chitosananion-exchange membranes for protein separations. J. Membr. Sci.1998, 148 (2), 195−205.(5) Popuri, S. R.; Vijaya, Y.; Boddu, V. M.; Abburi, K. Adsorptiveremoval of copper and nickel ions from water using chitosan coatedPVC beads. Bioresour. Technol. 2009, 100 (1), 194−199.(6) Davis, T. A.; Volesky, B.; Mucci, A. A review of the biochemistryof heavy metal biosorption by brown algae. Water Res. 2003, 37 (18),4311−4330.(7) Kang, D. W.; Choi, H. R.; Kweon, D. K. Stability constants ofamidoximated chitosan-g-poly(acrylonitrile) copolymer for heavymetal ions. J. Appl. Polym. Sci. 1999, 73 (4), 469−476.(8) Hosny, W. M.; Hadi, A. K. A.; El-Saied, H.; Basta, A. H. Metalchelates with some cellulose derivatives. Part III. Synthesis andstructural chemistry of nickel (II) and copper (II) complexes withcarboxymethyl cellulose. Polym. Int. 1995, 37 (2), 93−96.(9) Saliba, R.; Gauthier, H.; Gauthier, R.; Petit-Ramel, M. Adsorptionof copper(II) and chromium(III) ions onto amidoximated cellulose. J.Appl. Polym. Sci. 2000, 75 (13), 1624−1631.(10) Chan, W.-C.; Wu, J.-Y. Dynamic adsorption behaviors betweenCu2+ ion and water-insoluble amphoteric starch in aqueous solutions.J. Appl. Polym. Sci. 2001, 81 (12), 2849−2855.(11) Ngah, W. S. W.; Fatinathan, S. Adsorption of Cu(II) ions inaqueous solution using chitosan beads, chitosan-GLA beads andchitosan-alginate beads. Chem. Eng. J. 2008, 143 (1−3), 62−72.(12) Xu, S.-M.; Feng, S.; Yue, F.; Wang, J.-D. Adsorption of Cu(II)ions from an aqueous solution by crosslinked amphoteric starch. J.Appl. Polym. Sci. 2004, 92 (2), 728−732.(13) Yang, Y.; Shao, J. Synthesis of sulfhydryl chitin and itsadsorption properties for heavy metal ions. J. Appl. Polym. Sci. 2000, 77(1), 151−155.(14) Fourest, E.; Volesky, B. Alginate properties and heavy metalbiosorption by marine algae. Appl. Biochem. Biotechnol. 1997, 67 (3),215−226.(15) Yang, Y.; Wei, X.; Wan, J.; Meng, Z. Equilibrium and kineticcharacteristic of adsorption of Cu2+, Pb2+ on a novel anionic starchmicrospheres. Water, Air, Soil Pollut. 2011, 219 (1−4), 103−112.(16) Kumar, R.; Liu, D.; Zhang, L. Advances in proteinousbiomaterials. J. Biobased Mater. Bioenergy 2008, 2 (1), 1−24.(17) Liu, D.; Tian, H.; Kumar, R.; Zhang, L. Self-assembly of nanohydroxyapatite or aragonite induced by molecular recognition to soyglobulin 7S or 11S. Macromol. Rapid Commun. 2009, 30 (17), 1498−1503.(18) Liu, D.; Chen, H.; Chang, P. R.; Wu, Q.; Li, K.; Guan, L.Biomimetic soy protein nanocomposites with calcium carbonatecrystalline arrays for use as wood adhesive. Bioresour. Technol. 2010,101 (15), 6235−6241.(19) Wang, F.; Li, P.; Yang, G.; Cui, Y. Capabilities for chelatingmetals by modified soy protein gel. Technol. Water Treatment 2007, 33(12), 38−41.(20) Hwang, D.; Damodaran, S. Metal-chelating properties andbiodegradability of an ethylenediaminetetraacetic acid dianhydridemodified soy protein hydrogel. J. Appl. Polym. Sci. 1997, 64 (5), 891−901.(21) Soares, F.; Farina, M.; Santos, F.; Souza, D.; Rocha, J.; Nogueira,C. Interaction between metals and chelating agents affects glutamatebinding on brain synaptic membranes. Neurochem. Res. 2003, 28 (12),1859−1865.(22) Deng, L.; Su, Y.; Su, H.; Wang, X.; Zhu, X. Sorption anddesorption of lead (II) from wastewater by green algae Cladophorafascicularis. J. Hazard. Mater. 2007, 143 (1−2), 220−225.(23) Maruca, R.; Suder, B. J.; Wightman, J. P. Interaction of heavymetals with chitin and chitosan. III. Chromium. J. Appl. Polym. Sci.1982, 27 (12), 4827−4837.



mailto:[email protected]

(24) Juang, R.-S.; Shao, H.-J. A simplified equilibrium model forsorption of heavy metal ions from aqueous solutions on chitosan.Water Res. 2002, 36 (12), 2999−3008.(25) Dzul Erosa, M. S.; Saucedo Medina, T. I.; Navarro Mendoza, R.;Avila Rodriguez, M.; Guibal, E. Cadmium sorption on chitosansorbents: Kinetic and equilibrium studies. Hydrometallurgy 2001, 61(3), 157−167.(26) Nagib, S.; Inoue, K.; Yamaguchi, T.; Tamaru, T. Recovery of Nifrom a large excess of Al generated from spent hydrodesulfurizationcatalyst using picolylamine type chelating resin and complexane typesof chemically modified chitosan. Hydrometallurgy 1999, 51 (1), 73−85.(27) Eiden, C. A.; Jewell, C. A.; Wightman, J. P. Interaction of leadand chromium with chitin and chitosan. J. Appl. Polym. Sci. 1980, 25(8), 1587−1599.(28) Kamari, A.; Pulford, I. D.; Hargreaves, J. S. J. Binding of heavymetal contaminants onto chitosans − An evaluation for remediation ofmetal contaminated soil and water. J. Environ. Manage. 2011, 92 (10),2675−2682.(29) Langmuir, I. The constitution and fundamental properties ofsolids and liquids. Part I. Solids. J. Am. Chem. Soc. 1916, 38 (11),2221−2295.(30) Ho, Y.-S. Removal of copper ions from aqueous solution by treefern. Water Res. 2003, 37 (10), 2323−2330.(31) McKay, G.; Blair, H. S.; Gardner, J. R. Adsorption of dyes onchitin. I. Equilibrium studies. J. Appl. Polym. Sci. 1982, 27 (8), 3043−3057.(32) Bhattacharyya, K. G.; Sharma, A. Azadirachta indica leaf powderas an effective biosorbent for dyes: A case study with aqueous CongoRed solutions. J. Environ. Manage. 2004, 71 (3), 217−229.(33) Malash, G. F.; El-Khaiary, M. I. Piecewise linear regression: Astatistical method for the analysis of experimental adsorption data bythe intraparticle-diffusion models. Chem. Eng. J. 2010, 163 (3), 256−263.(34) Chien, S. H.; Clayton, W. R. Application of Elovich equation tothe kinetics of phosphate release and sorption in soils 1. Soil Sci. Soc.Am. J. 1980, 44 (2), 265−268.



Documents

Adsorption Behavior of Heavy Metal Ions from Aqueous Solution by